This Time, It’s Personal
Original source here
Overcoming lung cancer treatment resistance will require predictive biomarkers that take into account significant patient variability
Since the introduction of immunotherapy almost a decade ago and the development of new targeted therapies, we have seen significant progress in lung cancer management, particularly non-small cell lung cancer (NSCLC). This progress is, in part, thanks to the introduction of precision medicine solutions that allow the selective use of targeted therapies. Nonetheless, response rates for metastatic NSCLC still barely reach 30 percent – meaning that, on average, only around three of every 10 patients will benefit from treatment over time. Treatment resistance is still a huge problem and, though commendable progress has been made, we are still far from realizing true success.
Statistically, there are about 230,000 new cases of NSCLC in the US each year, resulting in around 120,000 deaths. When considering targeted therapies, we look for specific driver mutations to decide whether we should use specific drugs, but this is relevant to only about 14 percent of patients. For immunotherapy, which is relevant for 85 percent of patients, the situation is far worse. We have some biomarkers that we use mainly as prognostic tools, but this is a very limited arsenal to test for resistance and the accuracy of these biomarkers is unimpressive. The bottom line? For most patients, there are no good biomarkers.
Patients enter treatment with multiple therapeutic options – but, if we don’t have a way to identify which patients will benefit from specific options, all we can do is start treatment and hope for the best. Almost by definition, we will waste time for some (if not most) of our patients by having to treat them with therapies that may not work. Therefore, if we can discover biomarkers that tell us beforehand what the right treatment might be and what response trajectories might look like, clinicians can make informed decisions, identify resistance, intervene sooner, and choose next-line therapies based on patient and cancer biology, rather than relying on one-size-fits-all protocols.
There are currently hundreds of clinical trials investigating different combinations of lung cancer treatments. If even a fraction of these are successful, two to three years from now, clinicians could have the choice of 10–15 potential combinations of first-line treatments. But, to choose appropriately, we desperately need appropriate biomarkers to support clinical decision-making in first-line treatment and to help identify the best combination for each patient.
Standing in the way
Cancer is a very complex disease. There is a continuous interaction between the patient, tumor, and therapy – so, during biomarker discovery, we need to consider a complex dynamic system that differs both between patients and within the same patient and changes over time. This is not ideal when trying to identify and develop robust biomarkers. Access to tissue is also a challenge; tumor tissue is not always available or usable, limiting lab professionals’ ability to look for new biomarkers.
Although biomarker development is far from simple, we shouldn’t shy away from its complexity. Instead, we should broaden our understanding of cancer biology and tumor–patient–therapy interplay. We should use new bioinformatics and machine learning tools that can make sense out of all the signals, patterns, and markers that play a role in disease dynamics. And we should understand that cancer is not just one snapshot in time; rather, we should think about it as an almost continuous process and try to divide our biomarker search into different stages of the disease – before, during, and immediately after treatment – for continuous monitoring.
The earlier the better
I believe the two areas that require the most focus and investment right now are early detection and controlled or guided treatment planning to help clinicians make more informed decisions. There is a huge need in this area because we must gather a holistic picture of our patients over time. We are not just looking for snapshots of the disease; we are building a dynamic monitoring system. Early detection is one of the most important efforts we can make in cancer management because the earlier you catch it, the higher the chance of better outcomes for the patient. Many companies are focusing resources in this area, but most patients (especially those with lung or ovarian cancers) are still diagnosed late in their journey. In these cases, the tumor is already metastatic and the focus shifts from early detection to optimizing the patient’s treatment plan.
Where to next?
Right now, biomarker testing is going in all different directions. Many companies are trying to innovate in this area; that’s good news but having so many different players may also make it difficult to find one good solution. The future of biomarker testing heavily depends on pharmaceutical companies’ approach. If they continue to search for cancer drugs to treat “all-comers” populations – ignoring the fact that there is huge variability between patients – we will see more drugs doing excellent work, but for only a small percentage of patients.
We understand that we won’t find the be-all and end-all answers in DNA – genetics and genomics won’t provide us with the “silver bullet” we are looking for – but we do understand the issue of sheer complexity that cancer brings. That’s why companies are now attempting to combine biomarkers, bringing together genomics, proteomics, the immune system, and the microbiome to provide a deeper understanding of disease dynamics. By combining biomarkers, disease monitoring approaches, multi-omics, and AI-based systems that generate insights beyond what human pathologists can see under a microscope, we can revolutionize the field of precision oncology and do better for our patients – now and in the future.